1. Field of the Invention
The present invention relates to a TFCI decoding circuit and decoding method for decoding a TFCI (Transport Format Combination Indicator) codeword that is defined in the 3GPP (3rd Generation Partnership Project) as the third generation mobile communication system standard, and determining a TFI (Transport Format Indicator).
2. Description of the Prior Art
At present, in a communication system for portable telephones, a method for transmitting the information for a plurality of services with voice or packet multiplexed at multiple transmission rates within the same radio frame has been put into practical use. As means for indicating the content of multiplexed information or the information rate, a method for adding an information bit called a TFCI (Transport Format Combination Indicator) is provided and adopted in the 3GPP (3rd Generation Partnership Project) to create the third generation mobile communication system standard.
In accordance with the 3GPP standard, different services with packet or voice are allocated to a channel called a transport channel, and the data is encoded in a unit of transmission time called a TTI (Transmission Time Interval) for each transport channel, multiplexed into a physical channel (radio frame) and transmitted. Herein, the data amount per TTI on each transport channel can be judged based on an index called a TFI (Transport Format Indicator), in which the information called the TFCI represents a combination of TFIs on the multiplexed transport channel. Hence, if the TFCI is not normally received, the transmission rate for all the transport channels that are multiplexed can not be judged to disable the communication. Therefore, means for determining the TFCI at a low error rate is important. A relationship between the TFCI and the TFI is shown in FIG. 10.
FIG. 10 shows an example in which two transport channels 1 and 2 are multiplexed and the number of TFIs is 2. In this case, the TFCI has four combinations of TFIs.
One example of the conventional TFCI decoding circuit is described in Japanese Patent Laid-Open No. 2002-026735. The TFCI is converted into a TFCI codeword, and transmitted on a radio circuit. This TFCI codeword has a redundant bit length to withstand an error, whereby various methods for detecting and decoding the TFCI have been proposed. A relationship between the TFCI and the TFCI codeword is shown in FIG. 11. In this example of FIG. 11, the Hamming distance between the four codewords is about 16, with an equal error probability for other codewords is equal, as shown in FIG. 12.
However, there was a following problem with the prior art. In the 3GPP standard, TTI is different for each transport channel, whereby there is the possibility that the TFCI codeword is changed in a unit of minimum TTI. Therefore, in the conventional TFCI decoding circuit, the TFCI codeword was decoded in a unit of minimum TTI. For example, when the TTI is equal to 40 ms for all the transport channels, the TFCI codeword for each radio frame of 10 ms is added in common mode for four frames, whereby the signal-to-noise ratio is improved four times, with a lower error rate, as shown in FIG. 13. However, when a transport channel with TTI=20 ms and a transport channel with TTI=10 ms are multiplexed, as shown in FIG. 14, or when a transport channel with TTI=40 ms and a transport channel with TTI=10 ms are multiplexed, as shown in FIG. 15, there is the possibility that the TFCI is changed at every 10 ms, whereby the common mode addition is disabled in the maximum TTI (40 ms in FIG. 14 and 20 ms in FIG. 15). Therefore, the TFI information to be transmitted repetitively can not be effectively utilized. If the TTI is longer, the interleave (data rearrangement) period can be lengthened, whereby the error correcting ability for the transmission data itself is improved, but no TFCI error is corrected.
The conventional TFCI decoding circuit had a problem that because the TFCI codeword in a unit of minimum TTI was only employed to determine the TFI, the receiving performance was possibly worse due to an erroneous decoding.